What are your approaches to characterizing defects?

Hi, this is my first post here and I am experimenting with the different post types. This is an "article", which seems similar to text threads on other websites.

I am the design owner for a few metal parts. A rather constant issue is production documenting small defects (nicks, scratches, etc) on the production line. The parts aren't cheap, and they're industrial so often they come to me for evaluation. I have it pretty easy, my predecessors have done a few studies both with empirical specimens and ansys simulations, and I have a copy of peterson's.
Is there anything I'm missing?